Both type 1 and 2 diabetes are caused by inadequate insulin secretion, and in both diseases this can be ascribed in large part to loss of pancreatic beta cells that produce insulin. Both diseases can be cured by replacement of pancreatic beta cells by pancreas transplantation but this requires life long immunosuppression and there are far fewer organ donors than people with diabetes. An alternative approach to restoring pancreatic beta cells is to foster regeneration of beta cells in the patient. Is this plausible? In rodents yes, but human beta cell biology is clearly different to that in rodents. The studies proposed in this application would address the question, to what extent is beta cell regeneration feasible in humans? This goal will be approached by addressing four Specific Aims. First, we will establish the extent (and by what mechanisms) beta cell mass increases in human childhood. Second, we will establish the extent (and by what mechanisms) beta cell mass increases in response to obesity in humans. Third, we will establish if beta cell mass is maintained in non diabetic aging humans, and if so by what mechanisms. Fourth, we examine the ability of cord blood stem cells to transdifferentiate into beta cells in humans (since this approach is already in clinical trials). To accomplish these goals we will obtain well preserved human pancreas at autopsy in patients in who a recent ambulatory general medical examination (and labs) is available. We will apply a recently developed model to compute beta cell turnover permitting us to quantify beta cell turnover, and distinguish between new beta cell formation from beta cell replication and replication independent sources. These studies would be undertaken by a consortium with a long established track record of productive research with these resources.